[U-04] Geospatial Applications for Societal Benefits

[U04-05] Biomass Estimation of Rice Plants under Different Nitrogen Fertilizer Treatments based on spectral Indices

*San Lin Phyo¹, Yukihiro Takahashi¹ (1.Hokkaido University)

Keywords:Biomass of Rice Plants, Spectral Indices, Normalized Difference Spectral Index (NDSI), Spectrometer, Correlation and Regression

Rice is a major food crop around the world because it feeds over 50% of the population worldwide [FAO, 2019]. The total harvested area of rice is approximately 160 million hectares, producing more than 700 million tons annually [FAO, 2020]. The Food and Agriculture Organization (FAO) Rice Market Monitor (2018) reported that 681 million tons of rice were grown in Asia, representing 90% of the global rice production. According to OECD-FAO, 2018, rice production needs to increase approximately 40% by the end of 2030 as the global population will reach 9 billion by 2050. Therefore, [Zhao et al.,2013] mentioned that timely monitoring of rice growth status is crucial for global food security and agricultural sustainability. Biomass of rice plants is an important indicator for growth status monitoring and yield prediction. Crop biomass defined as the averaged dry weight per unit area is an important agronomic trait linked to plant genetics, growth rate, and productivity. It is also a key ecological indicator of light use efficiency and carbon stocks in agroecosystems [Li W et al.,2016]. Moreover, [Zou et al., 2002; tang et al., 2016] proved that a significant amount of CO2 is stored by plant in paddy fields. Therefore, if the biomass of rice plant can be estimated, carbon storage of rice plants may be predicted. Biomass of rice fields is changing at each growing stage and can vary depending on nitrogen fertilizer application. Although several research regarding biomass estimation of rice fields utilizing remote sensing have been published, there are still limitations and unknown information regarding biomass prediction. Therefore, the purpose of this research is to estimate temporal variation of biomass for rice plants based on spectral data. In this study, rice plants under four different nitrogen fertilizer conditions were used for biomass estimation. Firstly, the reflectance of rice plant was measured at each growth stage using spectrometer and then calculated Normalized Difference Spectral Index (NDSI). Based on spectral indices, the models for biomass were established by applying correlation and linear regression methods. After that, biomass of rice plants was predicted using the models created and the accuracy between predicted biomass and actual data was examined. In this seminar, the methods that were used for this study will be presented and the analysis results will be discussed.